A speech encoder is typically used to process noisy speech and tested using a moderate level of noise. Since substantial background noises are common in speech communications, the speech encoder may include its own “native” noise suppressor to attempt to suppress these background noises before the speech is encoded by a speech encoder. The speech encoder's noise suppressor may simply classify audio signals as stationary and non-stationary, (i.e., the stationary signal corresponding to noise and the non-stationary signal corresponding to speech). In addition, the speech encoder's noise suppressor is typically monaural, further limiting the classification effectiveness of the noise suppressor.